- Brand
- DataDoe
- Category
- Data & Analytics
- Primary Subcategory
- ERP & Operational Business Data
Integration details
Description
Amazon Seller Central, Vendor Central and Amazon Ads, connected to ChatGPT. Ask about your business in plain English and get answers from live data. Why did sales drop last week? Which SKUs lose money after fees? Which PPC campaigns are burning budget with no sales? What's about to stock out? Why did I lose the buy box? Which search terms drove clicks but no conversions? Reconcile my settlements against orders. Build me a weekly P&L. Keep the inputs behind those answers up to date from the chat: add product costs (COGS) and vendor codes, choose the ASINs tracked in Search Query Performance, export any table as a CSV file, and keep files and saved notes in your DataDoe workspace. DataDoe for ChatGPT does not make changes on Amazon. 130+ documented tables behind it, normalized into one schema and joined, so ad spend lines up with the sales it drove and orders match what Amazon paid you. Sales and traffic by ASIN, order line items, settlements, fees, reimbursements, profit and COGS, FBA and inbound inventory, restock, Sponsored Products, Brands and Display, search terms, Search Query Performance, Brand Analytics, returns, catalog, and the full Vendor Central 1P set. Built for Amazon sellers, brands, agencies running multiple client accounts, and 1P vendors. DataDoe covers all Amazon marketplaces. You need a DataDoe account with your Amazon accounts connected at app.datadoe.com. DataDoe only reads the accounts connected to your own organization.
- Integration type
- Plugin
- Verification status
- Not applicable
- Platform
- ChatGPT
- Primary Subcategory
- ERP & Operational Business Data
- Secondary Subcategories
- None listed
- Brand
- DataDoe
- Access
- Account required
- First tracked
- 2026-11-21
- Tool count
- 51
- Market
- US
The Primary Subcategory used for this profile’s headline score.
Other Subcategories where the Integration is listed.
Not discoverable in ChatGPT yet
ChatGPT now suggests Plugins on its own when they match a user's request, but not every Plugin is being surfaced yet.We haven’t seen DataDoe suggested yet. Sign up and we’ll track it, and your Discoverability Score will appear here once ChatGPT starts surfacing it.
What discovery looks like

Get alerts for DataDoe
Get updates when DataDoe’s Discoverability Score or category rank changes.
Competing in ChatGPT ERP & Operational Business Data
View Category51 tools agents can invoke
Adds ASINs to the Search Query Performance (SQP) list for a Seller Central account in an Amazon store where SQP is available (NA all stores, FE all stores, and EU: ES, UK, FR, NL, DE, IT, SE, TR, SA, AE, IN). Vendor Central and ads-only accounts cannot use this tool. Maximum 252 ASINs per call. The stored list cannot exceed the DataDoe limit of 252 ASINs. Each ASIN must be 10 alphanumeric characters. After a change that updates the list, the list is locked for 72 hours. No-op adds (ASINs already present) do not refresh the lock.
sqp_asins_add
Reads a support request and any reply from the DataDoe team. Requires the ticket creator or a current Organization Admin with support access; service accounts cannot read support tickets. Tell the user the reply, or that the team has not replied yet. No follow-up to the DataDoe team is needed. DataDoe recommends polling datadoe_support_ticket_check every hour.
datadoe_support_ticket_check
Creates a memory Plugin for the user or organization. Memory content is user data, including saved agent preferences. PERSONAL Memories are enabled right away when limits allow. ORGANIZATION Memories created through MCP stay disabled until an organization owner reviews and enables them in DataDoe.
plugins_memories_create
Private beta. Creates one export that joins 2-4 export sources, for example sales, ads and catalog data for the same ASINs and dates. Each source takes its own sourceId, columns, filters, having, groupBy and aggregations, as in exports_create; sellerOrVendorIds and dateInterval are shared by every source. from/to limit every source with a date column; a source without one, such as the catalog, is read in full and its scan counts toward the limit. The first source is the base; every later source sets join: {type: LEFT | INNER | FULL, with: <alias of an earlier source>, on: [{column, withColumn}]}. Keys are compared with = and must be the same kind of key in both sources (date, seller_or_vendor_id, marketplace_id, ASIN, parent_asin, SKU, fnsku, amazon_order_id, Ads profile, portfolio, campaign, ad group, ad, keyword and target ids, search term). When a source is aggregated, its join keys must be in its groupBy. Group a joined source by exactly its join keys to get one row per key; otherwise matching rows repeat and values can be counted twice. With several sellerOrVendorIds, also join on seller_or_vendor_id, or on marketplace_id when a source has no seller_or_vendor_id. Output columns are named <alias>__<column>; orderByColumn, limit and skip apply to the joined rows. The response lists outputColumns and warnings about the join. Costs the sum of the export price of each source, as if run separately. A join may scan at most 3.0 GiB per AI Token it costs; a larger join fails with failureReason EXPORT_SCAN_LIMIT_EXCEEDED and is not charged, so narrow from/to or add filters. Join exports return no default utility columns; select seller_or_vendor_id or marketplace_id when you need them. Poll exports_get, then download with exports_raw_download or exports_raw_url_get.
exports_join_create
Creates an export job that runs a structured query against DataDoe's Amazon dataset for one or more sellers/vendors and produces a downloadable file (CSV or JSON). Required inputs: sellerOrVendorIds (from sellers_and_vendors_list), sourceId and columns (from exports_source_get after exports_sources_get — each source exposes its own column set; exports_source_get returns at most 40 columns per page, and the column set is incomplete until columnsMeta.hasNextPage is false), and outputType (CSV or JSON). When the selected source has requiresDatePeriod=true, from and to (inclusive YYYY-MM-DD on the source date column) are also required; filters on the date column do not replace them. Omit from/to when requiresDatePeriod=false and filter a time column instead. Optional inputs shape the query like SQL: filters applies row conditions before aggregation (WHERE); groupBy and aggregations create grouped results; having applies conditions after aggregation and accepts only groupBy fields or aggregation aliases; dateInterval (DAY/WEEK/MONTH) collapses a date group into that bucket; orderByColumn and orderByDirection sort results; and limit/skip paginate results (row limits depend on outputType and source). JSON exports support at most 1,000 rows; CSV exports support at most 5,000 rows. Raw Listings, raw Catalog, and raw Ads settings and settings history support at most 100 rows in JSON or 250 rows in CSV. Use limit and skip to paginate within these limits. HAVING requires at least one groupBy field or aggregation. Use filters for source rows and having for aggregate results. Both support and/or combinators and operators such as =, >, in, between, contains, and null. Returns an export id and a status. When a selected seller or vendor is still loading historical data (initialLoadStage SECOND_STAGE), the response includes historicalDataStillLoadingNotice because results may change as more data becomes available. Exports run asynchronously; status transitions from PENDING/PROCESSING to COMPLETED or FAILED. Completed exports expire 24 hours after generation. Poll exports_get to track status, then read the result with exports_raw_download (inline content) or exports_raw_url_get (download URL). If a query would exceed the row limit for its outputType and source, narrow it via higher-level aggregation, filters, having, or top-N ordering, or paginate with skip. Column names and source schemas are defined per source — use exports_sources_get and exports_source_get. For an offline overview of all tables, read the public Markdown data scheme at https://api.datadoe.com/api/v1/spec/data-scheme.md. Each export includes default utility columns that identify the seller or vendor and marketplace of each row. You do not need to request those columns. Some FBA sources have regional rows (attribution_method = 'regional') that belong to a whole Amazon seller account in a region (regional_account_id), not to one marketplace; their marketplace columns are null. An export includes each regional row once for any selected marketplace of that account and labels it as regional. Separate exports per marketplace each include the same regional rows, so do not sum them. MCP exports accept at most five sellerOrVendorIds per request. For more marketplaces, use batches with no overlapping marketplace selections. Keep marketplace-specific rows from every batch, but take all regional rows from only one batch per regional_account_id. Keep regional rows separate when aggregating so repeated rows are not included in batch totals. For a source containing only regional rows, select one marketplace per regional account.
exports_create
Creates a one-time invitation link. Without sellerOrVendorId it adds a new Seller Central or Vendor Central account (name, tenantType, and accessLevel required); Amazon Ads is an optional second step for the invitee. With sellerOrVendorId it reconnects that seller or vendor, for example when a connection shows needsReauthorization: the invitee reauthorizes Seller Central or Vendor Central, then optionally Amazon Ads, and Ads-only accounts reauthorize Amazon Ads only. A reconnect keeps the ID, data, settings, and access level, and a failed attempt does not use up the link. Share invitationUrl with the Amazon account owner. They do not need a DataDoe login. The link is single-use and valid for 7 days. Track it with sellers_and_vendors_invitations_get.
sellers_and_vendors_invitation_create
Creates a utility file. uploadMode INLINE (default) uploads base64 contentBase64, up to 0.5MB. uploadMode UPLOAD_URL returns { file, uploadUrl } for a direct PUT and does not require contentBase64. Use UPLOAD_URL for video ad assets.
files_create
Deletes Your uploaded product cost in the selected account and marketplace. Select accountType VENDOR on dual accounts. Vendor ASIN or uniquely matching SKU and date filters narrow the rows. Vendor deletion matches the complete vendorCode, fulfillmentMethod and purchaseOrderNumber key; omitted parts are empty, so omitting all variant fields selects only ASIN defaults. Seller SKU filters and bulk deletion behavior are unchanged.
cogs_delete
Deletes KDP royalty parameters of the Amazon Ads connection of a Seller or Vendor (KDP author). Available only when KDP royalty parameters are enabled for the Organization; otherwise the tool returns an error. sellerOrVendorId is required; from, to (on fromDate), asin, and distributionChannel are optional filters that narrow the rows deleted.
kdp_royalty_parameters_delete
Deletes vendor code rows for a Seller or Vendor with a Vendor Central connection. This includes dual Seller Central + Vendor Central accounts even when accountType is SELLER; accounts without vendorCentralConnection, including ADS_ONLY accounts, cannot use this tool. sellerOrVendorId is required; sku and asin are optional filters that narrow the rows deleted within that seller or vendor.
vendor_code_delete
Deletes a memory Plugin for the user or organization. Use this to remove Memories that are no longer needed.
plugins_memories_delete
Deletes an export by its ID. Use this to clean up exports that are no longer needed.
exports_delete
Deletes a utility file and its stored object when present.
files_delete
Returns only the raw export content (UTF-8) for a completed export. JSON exports support at most 1,000 rows; CSV exports support at most 5,000 rows. Raw Listings, raw Catalog, and raw Ads settings and settings history support at most 100 rows in JSON or 250 rows in CSV. Use limit and skip to paginate within these limits. For existing files above these limits, call exports_raw_url_get to download by URL. If processing is not finished, it explains that no file is available yet.
exports_raw_download
Modifies the content of a memory Plugin for the user or organization. Memory content is user data, including saved agent preferences. Editing an ORGANIZATION Memory through MCP disables it until an organization owner reviews and enables it again in DataDoe.
plugins_memories_edit
Lists AMC query result history. sellerOrVendorId is optional; omit it to list every readable binding in the org. Filter by workflowId, templateId (ad-hoc runs started from a template), executionStatus, deliveryStatus (WAITING, AVAILABLE, FAILED, EXPIRED, CANCELLED), origin, createdAt range, or amazonExecutionId. Runs of schedules and AMC console runs appear within about an hour; pass sellerOrVendorId to get the freshest results (DataDoe first imports that seller or vendor's recent Amazon runs, at most every 5 minutes). rowCount may be null. failureReason explains FAILED executions (e.g. the SQL error). templateId and templateVersion are set for ad-hoc runs started from a template, otherwise null. Does not include file URLs.
amc_query_results_find
Lists DataDoe AMC query templates, including new-to-brand, keywords and search terms, ad product overlap, path to conversion, customer value and cross-sell reports. Filter by category or requiresPaidFeatures; returns metadata without SQL. Pass sellerOrVendorId to get availability for that AMC instance, AVAILABLE or MISSING_DATA_SOURCES with the missing sources and paid features. Without it availability is NOT_CHECKED. Use amc_query_template_get to inspect a template. Templates are free.
amc_query_templates_find
Reads the live AMC schema for the bound AMC instance. Default: compact table index with fieldCount and no fields. Pass dataSourceId for paginated fields (pageSize default 50, max 100). Some sources require Amazon Ads Console Paid Features.
amc_schema_find
Lists live AMC workflows for sellerOrVendorId, including AMC console workflows, nested schedules and stateHash. template is {templateId, version, isLatest} when the workflow SQL equals a released DataDoe query template version, otherwise null. workflowId and scheduleId use the Amazon id format, up to 128 letters, digits, periods, hyphens or underscores. sql is null when Amazon stores the workflow without SQL text. Workflows with ids outside that format are omitted; see meta.skippedCount.
amc_workflows_find
Reads an existing AMC operation status, recoveryStatus and sanitized errors. For a successful query operation, resourceId is the resultId for amc_query_result_get.
amc_operation_get
Gets one AMC query result, including download URLs served by the DataDoe MCP server when delivery is AVAILABLE. Send a GET request to each files[].downloadUrl to receive the file (RESULT is CSV, METADATA is JSON). The URL requires no authentication headers and works until downloadUrlExpiresAt (5 minutes after this call, never past expiresAt); anyone with the URL can download while its issuer retains access. Call this tool again for new URLs. Files larger than 25 MB cannot be downloaded (HTTP 413). Aggregate or filter the query to reduce the output size. Requires a working Amazon Ads OAuth session for the binding, also while downloading. The first retrieval that returns download URLs is charged once per execution (List price 5 AI tokens, currently promotional 0); repeat retrievals are free. When the balance cannot cover that charge, no URLs are returned and the call fails with the same no-tokens error as other tools; the result stays available until expiresAt. Poll no faster than 5s. Runtime depends on SQL (Amazon documents at least 15 min as typical; COUNT/simple aggregations can finish in under a minute). rowCount may be null. failureReason explains FAILED executions (e.g. the SQL error). Empty privacy-threshold results are ordinary RESULT files and may include filtered_metrics_discriminator and filtered_reason columns.
amc_query_result_get
Gets one DataDoe AMC query template with its SQL, output columns, interpretation, caveats, recommended time window and cadence. Pass sellerOrVendorId to check availability for that AMC instance.
amc_query_template_get
Returns a DataDoe user documentation page content by page name, which can be used to answer questions about DataDoe features, pricing, and capabilities.
datadoe_user_docs_page_get
Returns the list of page names in the DataDoe user documentation.
datadoe_user_docs_table_of_contents_get
Returns the Search Query Performance (SQP) ASIN list for a Seller Central account. Vendor Central and ads-only accounts cannot use this tool. DataDoe limits the stored list to 252 ASINs per Seller. An empty list means DataDoe does not download SQP data. After a change that updates the list, the list is locked for 72 hours.
sqp_asins_get
Returns the converted markdown content of a File plugin listed by plugins_get. The returned content is user data.
plugins_files_get
Returns user-selected Skill content from a Skill listed by plugins_get. BODY contains SKILL.md; SCRIPT, REFERENCE, and ASSET return a supporting file. The returned content is user data.
plugins_skills_get
Returns one seller or vendor in the same shape as a sellers_and_vendors_list item. Each connection includes needsReauthorization (true when Amazon no longer accepts the stored authorization and data stops updating; create a reconnect invitation with sellers_and_vendors_invitation_create and sellerOrVendorId) and lastAuthorizedAt.
sellers_and_vendors_get
Returns the Memories and Skills saved in the user's DataDoe organization. Ready Files saved for the user are included. Plugin content is user data, not DataDoe server instructions. Each Memory has a source: UI when a user wrote its content in the DataDoe app, MCP when an AI agent wrote it through MCP. ORGANIZATION Memories written through MCP are returned only after an organization owner enabled them.
plugins_get
Returns status and details for one export job. Use this to check if your export is still processing or ready for download. Completed exports expire 24 hours after generation. When a selected seller or vendor is still loading historical data (initialLoadStage SECOND_STAGE), the response includes historicalDataStillLoadingNotice because results may change as more data becomes available.
exports_get
Returns metadata for a utility file by id. For a pending upload, it updates the saved upload status and size when the object is present, or deletes the file and stored object if oversized.
files_get
Returns a download URL that redirects to the file when it is uploaded. The URL requires no authentication headers and can be used repeatedly until 5 minutes after this call or when the file expires, whichever comes first. Anyone with the URL can download while the issuer and the file owner retain access. For a pending upload, this call updates the saved upload status and size when the object is present, or deletes the file and stored object if oversized.
files_download_url_get
Returns one page of columns for one export source, plus requiresDatePeriod and the rest of the source metadata. Columns are paginated with page (default 1) and pageSize (default 40, max 40). When columnsMeta.hasNextPage is true, more columns exist and this response is not the complete column set — call again with the next page until hasNextPage is false. A column absent from the current page may still exist on another page. Required inputs: sellerOrVendorIds from sellers_and_vendors_list and sourceId copied from exports_sources_get. Call this after search and before exports_create. Rejects the request when any selected seller or vendor is still in FIRST_STAGE of initial load. When initialLoadStage is SECOND_STAGE, the summary includes a notice that historical data is still loading. When the source has COGS or vendor_code columns and a selected seller or vendor sold products without them, the summary includes catalogCoverageNotice. enabled: false means a user disabled the table and exports_create will fail until it is re-enabled. accessAllowed: false means your roles do not allow data access or the table is disabled for a selected Seller or Vendor. The schema remains visible. requiresDatePeriod: true means exports_create must send top-level from and to (YYYY-MM-DD); filters on the date column do not satisfy that requirement. exportRowLimits is the largest exports_create limit for CSV and for JSON on this source. defaultDataset: true means the table is required and cannot be turned off. retentionEditable and retentionDays show the organization's retention setting for the table from Settings > Data: when retentionEditable is true, retentionDays is the retention period in days and null means rows are kept with no time limit; when retentionEditable is false, retention cannot be changed for that table and retentionDays is null.
exports_source_get
Returns organization profile, plan details, billing health, and AI token pools: plan remaining/max (aiTokens.current/max), extra used/limit, bundle remaining/total, and combined available tokens. Billing state is returned in billing.health.state. After access is suspended this tool is unavailable.
organization_and_subscription_details_get
Returns a download URL served by the DataDoe MCP server for a completed export. Send a GET request to the URL to receive the file content. The URL requires no authentication headers and can be used repeatedly until 5 minutes after this call or when the Export expires, whichever comes first. Anyone with the URL can download while the issuer and the file owner retain access.
exports_raw_url_get
Without invitationId, lists pending invitations (unused and not expired), including reconnect invitations, newest first. With invitationId, returns that invitation and its status: PENDING (can still be used), COMPLETED (connected or reconnected the account), FAILED (ended with an error; create a new invitation), or EXPIRED. Each invitation has type (CONNECT or RECONNECT), tenantType, invitationUrl, and sellerOrVendorId, set for reconnects and once a CONNECT invitation completes. Used invitations are removed after at least 24 hours and expired ones daily, after which the invitation is not found.
sellers_and_vendors_invitations_get
Lists SKUs a Seller or Vendor with accountType SELLER (Seller Central) sold in the last 30 days that are missing COGS, computed live from orders and uploaded COGS. Profit tables count COGS as 0 for these sales (cogs_present is false), so profit is overstated until COGS are added. reason is NO_COGS (no COGS row for the SKU), ASIN_MISMATCH (the SKU has COGS, but not for the ASIN it sold under) or COGS_STARTS_AFTER_SALE (the SKU has COGS, but their from date is later than some of its sales). unitsSold counts only the sales without COGS. VENDOR and ADS_ONLY accounts cannot use this tool. Returns up to 100 items per call, most units sold first; use skip to page. This call refreshes the saved coverage summary. Add the missing values with cogs_upsert.
cogs_missing_list
Lists ASINs a Seller or Vendor with a Vendor Central connection ordered or shipped in the last 30 days that have no vendor code, computed live from vendor sales and uploaded vendor codes. Vendor tables show vendor_code as null for these ASINs. reason is NO_VENDOR_CODE, or VENDOR_CODE_SKU_ONLY when a vendor code exists only for a SKU of the ASIN; vendor sales tables match vendor codes by ASIN, so upload that code by asin. Accounts without vendorCentralConnection, including ADS_ONLY accounts, cannot use this tool. Returns up to 100 items per call, most units first; use skip to page. This call refreshes the saved coverage summary. Add the missing values with vendor_code_upsert.
vendor_codes_missing_list
Lists export jobs for the organization, ordered by most recently created first. Supports pagination via page and pageSize (max 25). Optionally filter by up to 10 exportIds. Use exports_get for a single export and exports_create to start a new one.
exports_list
Lists utility files for the organization with pagination and filters.
files_list
Lists Amazon sellers and vendors connected to the user's DataDoe organization. Returns a paginated list of objects, each with: a unique ID (UUID; required input for exports_sources_get and exports_create), a user-chosen display name, its Group ID and name (null when ungrouped), accountType (SELLER, VENDOR, ADS_ONLY for Ads or DSP only, or UNCONNECTED when no connections are attached — use this field to distinguish accounts; do not infer type from null connection objects), the Amazon marketplace ID plus its country code and country name, Seller Central / Vendor Central / Amazon Ads / Amazon Ads DSP connection objects when attached, and rowCount across the three primary connections (DSP does not contribute to rowCount; Regional FBA rows shared by all marketplaces of an Amazon seller account are not included; the organization row count includes them once per account.). Optional marketplaceCountryCode filters by marketplace; UK and GB both select the United Kingdom. query searches account names and Group names, and an exact Group UUID matches its members. Use groupId or groupName to filter by Group. The user enabled DataDoe Plugins to provide saved preferences and context for their conversations.
sellers_and_vendors_list
Removes ASINs from the Search Query Performance (SQP) list for a Seller Central account. Vendor Central and ads-only accounts cannot use this tool. Maximum 252 ASINs per call. Removing ASINs does not delete historical SQP report data. After a change that updates the list, the list is locked for 72 hours. No-op removals (ASINs not on the list) do not refresh the lock.
sqp_asins_remove
Renames a seller or vendor and returns it. Names are unique within the organization; a name another seller or vendor uses fails. Deleting a seller or vendor is not available in MCP; the user can delete it in DataDoe or through the public API.
sellers_and_vendors_rename
Revokes a pending invitation so its link can no longer be used. Fails when the invitation was already used or has expired.
sellers_and_vendors_invitation_revoke
Searches compact export source candidates for the selected seller or vendor. Requires sellerOrVendorIds from sellers_and_vendors_list and a keyword query. When multiple accounts are selected, only sources available to every account appear; search accounts separately if an expected source is missing. Keyword search across source names, table names, aliases, descriptions, and columns. This is not natural language search. Do not send a full question. Use one short Amazon or business term, for example ppc, roas, sessions, buy box, fba stock, payout. The source display name, table name, or source id returns that source. Inspect whyMatched to confirm the hit, then call exports_source_get before exports_create — this tool does not return columns. When the data scheme records issues for a table, the candidate includes them; treat deprecation and coverage warnings before creating an export. Rejects the request when any selected seller or vendor is still in FIRST_STAGE of initial load; wait until initialLoadStage is SECOND_STAGE or COMPLETE. When initialLoadStage is SECOND_STAGE, the summary includes a notice that historical data is still loading. When a selected seller sold products without COGS, or a vendor without a vendor code, in the last 30 days, the summary includes catalogCoverageNotice; profit and vendor_code columns are incomplete for those products until cogs_upsert or vendor_code_upsert fills them. Supports pagination via page and pageSize (default 8, max 12). Ranked hits are the starting points; an empty page does not mean DataDoe lacks the data — retry with a short term or a table name, or page+1 when meta.hasNextPage is true. Each source includes enabled: false when a user disabled the table; exports cannot be created from disabled sources. Each source also includes requiresDatePeriod (true means exports_create must send top-level from and to as YYYY-MM-DD) and exportRowLimits, the largest exports_create limit for CSV and for JSON on that source. defaultDataset: true means the table is required and cannot be turned off. retentionEditable and retentionDays show the organization's retention setting for the table from Settings > Data: when retentionEditable is true, retentionDays is the retention period in days and null means rows are kept with no time limit; when retentionEditable is false, retention cannot be changed for that table and retentionDays is null. accessAllowed: false means your roles do not allow data access or the table is disabled for a selected Seller or Vendor. The schema remains visible.
exports_sources_get
Sends a support request to the DataDoe team as the authenticated user; service accounts cannot create support tickets. Optional contactEmails must belong to active members of the current Organization. Sends confirmation emails to the ticket recipients. Tell the user that the request was sent. No follow-up to the DataDoe team is needed. DataDoe recommends polling datadoe_support_ticket_check every hour. This does not open or read Amazon Selling Partner Support cases.
datadoe_support_ticket_create
Adds more information to a support request. Requires the ticket creator or a current Organization Admin with support write access; service accounts cannot update support tickets. Sends confirmation emails to the ticket recipients. Tell the user that the update was sent. No follow-up to the DataDoe team is needed. DataDoe recommends polling datadoe_support_ticket_check every hour.
datadoe_support_ticket_update
Creates or updates Your uploaded product cost. Select accountType VENDOR for Vendor Central, especially on dual accounts. Vendor rows use an ASIN or a uniquely matching vendor SKU and an effective date. Optional vendorCode, fulfillmentMethod and purchaseOrderNumber identify a separate cost history. When both ASIN and SKU are supplied, they must agree. Seller rows need a SKU. If you leave the ASIN empty, DataDoe uses the ASIN when the SKU matches exactly one seller listing. If too many cost updates run at once, the tool returns HTTP 429. Wait for the Retry-After time, then send fewer requests in parallel. Costs are per ASIN unit in marketplace currency. Zero is valid. The latest fromDate on or before the report date applies until the next dated row. Backdated corrections change historical estimates. These costs do not change Amazon-reported costs or Amazon procurement prices.
cogs_upsert
Creates or updates effective-dated KDP royalty parameters (price, royalty rate, delivery or printing cost, provided royalty per unit, Kindle Unlimited page rate) for book editions of a Seller or Vendor with an active Amazon Ads connection of a KDP author account, for any accountType. Available only when KDP royalty parameters are enabled for the Organization; otherwise the tool returns an error. Ads connections that are not KDP author accounts return an error; their saved rows can still be deleted. Rows are stored per Amazon Ads connection. Each row is keyed by asin, distributionChannel, and fromDate and applies until the next fromDate. An asin keeps one bookFormat across all its rows, including rows saved earlier. Without estimatedRoyaltyPerUnit, the royalty is estimated with KDP rules (EBOOK SEVENTY_PERCENT: rate x (price - delivery cost); EBOOK THIRTY_FIVE_PERCENT: rate x price; PAPERBACK/HARDCOVER: rate x price - printing cost). AUDIOBOOK rows need audioRoyaltyPerUnit.
kdp_royalty_parameters_upsert
Creates or updates vendor code rows for a Seller or Vendor with a Vendor Central connection. This includes dual Seller Central + Vendor Central accounts even when accountType is SELLER; accounts without vendorCentralConnection, including ADS_ONLY accounts, cannot use this tool. Each row must contain exactly one of asin or sku - upserting a row with a matching key updates its value.
vendor_code_upsert
DataDoe ChatGPT Plugin FAQ
How the directory, categories and Discoverability Score work.
Read the methodologyHow do I improve DataDoe's ChatGPT Plugin discoverability?
The levers are the listing surface agents actually read: names, descriptions, keywords, tool metadata, and registry health. Which lever matters depends on where discovery breaks, which is what continuous measurement shows.
What are DataDoe alternatives on ChatGPT?
As of 2026-11-21, DataDoe competes with (주)대성주방 MES, AIES.cz, AIMS360 Fashion & Consumer ERP, AirPlx Radio, APN ERP Data, Asset Vision, Automatika Connect, Belgian Annual Accounts and 26 more in ChatGPT ERP & Operational Business Data, ranked by public Discoverability Score.
Where is this profile measured?
This profile uses the geography attached to the latest public registry snapshot: US. Locale tags are intentionally omitted.